| Episode | Status |
|---|---|
Bradley Tusk joins Sourcery to discuss what he learned working closely with Travis Kalanick during Uber’s early regulatory battles, and how those experiences shaped his decision to shut down his ventu...
Bradley Tusk discusses his transition from running a $140M venture fund to an equity-for-services model, sharing insights from working with Travis Kalanick at Uber and Mike Bloomberg. He critiques the current AI infrastructure spending boom as potentially driven more by short-term valuations than long-term fundamentals, while exploring the intersection of venture capital, regulation, and political strategy. Tusk offers candid perspectives on fund economics, the challenges of institutional fundraising, and why mid-sized VC funds struggle to generate meaningful returns.
Tusk questions whether the projected $2 trillion in AI infrastructure spending (data centers, GPUs, energy) is genuinely necessary or primarily designed to boost short-term share prices. He explores how DeepSeek's inference model could potentially reduce compute needs by 90%, making much of this investment wasteful, while acknowledging potential positive externalities like nuclear energy development.
Tusk argues that the fundamental choice between zero-sum thinking (Trump's worldview of accumulation and power) versus abundance mentality (relationships and purpose) determines both individual happiness and humanity's survival. He shares his personal happiness framework based on behavioral economics research and explains why he runs a money-losing bookstore instead of flying private.
Tusk recounts becoming Uber's first political advisor by accident in 2010, taking equity instead of fees, and developing the strategy of turning customers into political advocates. This experience shaped his entire investment philosophy and led to his current equity-for-services approach after realizing traditional VC fund economics didn't work for him.
Tusk explains the brutal economics of running a mid-sized venture fund, revealing why he shut down his $140M fund despite strong performance. He breaks down how management fees get consumed by operations, carry hurdles become nearly impossible to clear, and founders don't actually need the capital - they need the expertise.
Tusk shares specific examples of AI valuation disconnect, including a deepfake prevention company with $1M ARR that raised at $140M post-money (he offered $35M post). He explains why he's avoiding most AI deals despite the sector's promise, drawing parallels to his decision to skip cannabis investments based on Canadian market comps.
Tusk argues that despite the narrative of Silicon Valley-DC alignment, Trump is systematically undermining everything that makes the American economy great: IP protection, R&D, higher education, immigration, free trade, and independent institutions. He sees current policies as crony capitalism benefiting a few at everyone else's expense.
Tusk explains why AI's potential to create massive unemployment (potentially 18%) requires Universal Basic Income to prevent economic collapse. He argues that even if AI creates new industries long-term, the 17-25 year transition period will devastate consumer spending unless people have money, making UBI essential for capitalism to function.
Tusk offers an unapologetic defense of Travis Kalanick, calling him one of the two smartest people he's ever worked with (alongside Mike Bloomberg). He explains Travis's unique combination of intelligence, relentlessness, and vision, while acknowledging mistakes were primarily of omission - focusing too much on growth at expense of management.
What Travis Kalanick Taught Bradley Tusk, & Why He Closed His VC Fund
Ask me anything about this podcast episode...
Try asking: